paper

Asteroid sizes determined with thermophysical model and stellar occultations

arXiv:2505.09437 · doi:10.1051/0004-6361/202554476

Abstract

Context. The sizes of many asteroids, especially slowly rotating, low-amplitude targets, remain poorly constrained due to selection effects. These biases limit the availability of high-quality data, leaving size estimates reliant on spherical shape assumptions. Such approximations introduce significant uncertainties propagating, e.g. into density determinations or thermophysical and compositional studies, affecting our understanding of asteroid properties. Aims. This work targets poorly studied main-belt asteroids, most of which lacked shape models. Using only high-quality dense light curves, thermal IR observations (incl. WISE), and stellar occultations, we aimed to produce reliable shape models and scale them via two independent techniques, allowing size comparison. We conducted two campaigns to obtain dense photometric light curves and to acquire multi-chord stellar occultations. Methods. Shape and spin models were reconstructed using lightcurve inversion. Sizes were determined by (1) thermophysical modeling with the Convex Inversion Thermophysical Model (CITPM), optimizing spin and shape models to visible lightcurve and IR data, and (2) scaling shape models with stellar occultations. Results. We obtained precise sizes and shape models for 15 asteroids. CITPM- and occultation-derived sizes agree within 5% in most cases, demonstrating the modeling's reliability. Larger discrepancies usually stem from incomplete occultation chord coverage. The study also gives insights into surface properties incl. albedo, roughness and thermal inertia. Conclusions. Using high-quality data and an advanced TPM integrating thermal and visible data with shape adjustment enabled precise size estimates comparable to those from multi-chord stellar occultations. We resolved major inconsistencies in previous size estimates, providing solid input for future studies on asteroid densities and surfaces.

References in corpus (50)